In recent years, along with rapid advances of digital image compression (encoding) techniques represented by MPEG and JPEG, it has become possible to efficiently transmit and receive a high-quality video image with a small information amount in real time. Even in a case of transmission of image data, an advancement of a digital modulation and demodulation technique enables high-speed digital communications using a satellite or an optical fiber. In addition, even in storage media, a large-capacity hard disk, a large-capacity DVD, and the like have been developed, and environments for a digital image to be transmitted or stored are being provided.
Encoded data (bit stream, bit string) compressed at high efficiency is sensitive to an error, so the data is protected by an error correction technique against a transmission line error or a read error from a storage medium. However, an error rate cannot be completely reduced to zero, so an error remains in the data. When the encoded data including the error is to be decoded to a decoded image, there is a case where, even if the error is 1 bit, a decoded image is significantly influenced thereby. Therefore, in order to reduce the influence of the error on the decoded image, an error concealment technique for minimizing the error has been known and a method of replacing an error generation portion by a normal image or a gray image has been proposed (see, for example, Patent Document 1    Patent Document 1: JP 2004-528752 A (pages 14 to 17 and FIG. 3)